Multisensor Data Fusion

The group focuses on mathematical modelling and development of computational methods for challenging practical problems. Our scientific specialty is in estimation algorithms, and an application area of special expertise is positioning technology.

Our team is working on various positioning related projects. For the AUTOPORT project (https://autoport.fi) we develop algorithms for object detection, re-identification and tracking that rely on INS/GNSS measurements and camera images. We work with both model-based (e.g. Kalman-type filters) and data-based techniques (e.g. Artificial Neural Networks). We also study machine perception and predictive analysis to ensure the safety of all players in environments where autonomous vehicles operate alongside humans and non-autonomous vehicles.

In the GaitMaven project we collaborate with the University of Jyväskylä (Faculty of Sport and Health Sciences). Our teams are using small, inexpensive sensors to measure and analyze human movement in natural environments kinesiology research). At Tampere University we have been focusing on developing both hardware and software used in the project. Currently we focus on developing and testing various machine learning algorithms for estimating gait performance metrics and physiological parameters from multi-sensor data.

Furthermore, we are collaborating with multiple research groups and companies on various projects. In all these projects our team is responsible for analyzing sensor data and for developing algorithms that use these sensor measurements for solving various research questions.

Our educational objective is to train mathematical engineering experts for professional and academic careers.